I was sitting at my desk last Thursday evening, lamp on, tea going cold beside me, when I read a headline that stopped me mid-scroll. Johns Hopkins had published the results of a twenty-year follow-up to a brain training study, and the numbers were striking enough to make me set down my reading glasses and read the whole thing twice. Then I closed my laptop, walked to the kitchen, and stood there thinking about every advertisement I'd ever seen promising a sharper mind in fifteen minutes a day.
Because the truth about brain training and dementia is more complicated than any app store description wants you to believe. And more interesting.
The Study That Changed the Conversation
The study is called ACTIVE — Advanced Cognitive Training for Independent and Vital Elderly. The National Institutes of Health funded it. Researchers enrolled 2,802 adults aged 65 and older between 1998 and 1999 at six sites across the country, then randomly assigned them to one of three cognitive training groups or a control group that received no training at all.
Three training types were tested: memory strategies, reasoning exercises, and speed-of-processing tasks. Each group received eight to ten sessions, sixty to seventy-five minutes each, spread over five to six weeks. Some participants also received booster sessions at eleven and thirty-five months after the initial training.
In February 2026, the twenty-year follow-up data landed. Of the three training types, only one showed a connection to dementia risk. Speed-of-processing training (specifically, a visual task requiring participants to identify objects flashing on a screen in increasingly brief intervals) was associated with a 25 percent lower rate of dementia diagnosis compared to the control group. Forty percent of speed-training participants developed dementia over twenty years, versus 49 percent of the control group.
But here is the part most of the headlines buried. The reduction only appeared in participants who completed the booster sessions. People who did the initial five weeks of training and stopped? No measurable protection. The combination mattered: the specific type of exercise and the reinforcement over time.
Memory training showed no effect on dementia risk. Neither did reasoning training. Speed of processing, combined with booster sessions, was the only intervention that moved the needle. This is not a story about cognitive exercise in general — it is a story about one particular kind of practice, sustained over time, in one rigorous trial.
Nothing else in the cognitive training world has data like this. Not even close.
What the Apps Promise (and What the FTC Had to Say)
Open your phone's app store and search "brain training." You'll find dozens of options, most wrapped in cheerful colors and bold claims. Lumosity. Peak. Elevate. CogniFit. They range from free to $19.99 a month, and their marketing tends toward words like "scientifically designed" and "clinically proven."
Some of these apps are well-built. Genuinely fun, even. But fun and clinically proven are different things.
In January 2016, the Federal Trade Commission settled charges against Lumos Labs, the company behind Lumosity, for $2 million. The charge: deceptive advertising. Lumosity had claimed its games could help users perform better at work and school, reduce cognitive impairment from age-related conditions, and even protect against Alzheimer's disease. Jessica Rich, then director of the FTC's Bureau of Consumer Protection, put it plainly: "Lumosity preyed on consumers' fears about age-related cognitive decline. But Lumosity simply did not have the science to back up its ads."
That was a decade ago. The market has only grown since. And most apps still lean on vague references to "neuroscience" without citing specific clinical trials linking their product to real-world health outcomes. Getting better at a phone game is not the same thing as protecting your brain from dementia. Those are two very different claims, and the gap between them is where a lot of money changes hands.
The One App With Actual Clinical Evidence
There is one platform connected to the ACTIVE study results, and it deserves separate mention.
BrainHQ, made by Posit Science, offers the exercise called Double Decision, a direct descendant of the speed-of-processing task used in the ACTIVE trial. You see a vehicle flash in the center of the screen and a road sign in the periphery, both appearing for fractions of a second. The task adapts, getting faster and more cluttered as you improve.
BrainHQ costs $14 per month or $96 per year. A limited version is free. And this surprised me: some Medicare Advantage plans now include BrainHQ at no additional cost. As of 2025, the company cites more than 300 peer-reviewed studies showing benefits across cognition and mood as well as day-to-day functioning.
Ernestine, a retired teacher from Hendersonville who's been part of our Seasons of Grace circle for years, started using BrainHQ after I mentioned the ACTIVE study results one morning in March. She downloaded it on her iPad, spent twenty minutes with Double Decision, and called me that evening. "Eleanor, I couldn't catch a single road sign at first. Not one." She laughed about it. By week three, she was catching most of them. Whether that will translate into long-term cognitive protection, neither of us knows. The ACTIVE study used a controlled protocol with trained facilitators and specific booster schedules, not a self-directed app on a tablet. BrainHQ is the closest available version of what was studied, but closest and identical are not the same thing.
I want to be honest about the distinction because I think you deserve it.
What the Brain Actually Needs
Here is where I have to step back from brain training entirely and talk about what the research says our brains need most, because apps are a small piece of a much larger picture.
The 2024 Lancet Commission on dementia prevention identified fourteen modifiable risk factors that together account for roughly 45 percent of all dementia cases worldwide. Fourteen factors. Brain training apps are not among them. Isn't it interesting how the things with the strongest evidence are the ones nobody can sell you?
Here they are: low educational attainment in early life, hearing loss, high blood pressure, smoking, obesity, depression, physical inactivity, diabetes, excessive alcohol use, traumatic brain injury, air pollution, social isolation, untreated vision loss, and elevated LDL cholesterol in midlife.
Read that list slowly. Physical exercise. Treating hearing loss. Managing blood pressure. Staying connected to other people. These are the interventions with the strongest evidence base for reducing dementia risk. Not glamorous. Not downloadable. But backed by decades of population-level data involving hundreds of thousands of participants across multiple countries.
Actually, let me correct myself. The hearing loss piece is worth singling out because it's the one people overlook most. Untreated hearing loss is the single largest modifiable risk factor for dementia in midlife, according to the Lancet Commission. Hearing aids aren't just about hearing. They're about keeping the brain engaged with the world around it. A $3,000 pair of hearing aids with proper fitting may do more for cognitive health than a lifetime subscription to every brain training app combined!
And the Mediterranean diet keeps showing up in the data. A 2024 meta-analysis in GeroScience confirmed its association with reduced risk of cognitive impairment and Alzheimer's. Fruits, vegetables, whole grains, olive oil, fish. Less red meat. A July 2025 NPR report covered a large study finding that lifestyle changes after 60 (exercise combined with better diet and regular social engagement) produced cognitive scores equivalent to being one to two years younger. No supplement or screen involved. Just food, movement, and the people sitting across from you at dinner.
Our brains don't exist in isolation. They live inside bodies that need to move, eat well, sleep deeply, hear clearly, and feel connected. Any conversation about these apps that skips the basics is selling you the dessert and forgetting the meal.
The Difference Between Training and Living
My brother James called last Wednesday evening, same as always. He's seventy-two, lives in Savannah, retired from the post office six years ago. He told me he'd been doing the crossword in the Savannah Morning News every day and wanted to know if it counted as keeping his brain sharp.
"Counts as what?" I asked.
"You know. Keeping sharp. Warding off the decline."
I told him what I've told dozens of people at Seasons of Grace: doing crossword puzzles makes you better at crossword puzzles. It's enjoyable, it's engaging, and there is nothing wrong with it. But the evidence crosswords or Sudoku specifically prevent dementia is thin. What the research supports is novelty: learning something genuinely new, not repeating something you've already mastered. A new language, a musical instrument you've never played, a dance class where you have to remember steps you haven't learned yet.
"So my morning walk to the coffee shop counts more than the crossword?" James asked.
Probably. Yes.
What Medicare Covers (and What It Doesn't)
Medicare does not cover brain training apps. No subscription to BrainHQ, Lumosity, or any other platform is reimbursable under Part A or Part B.
What Medicare does cover: a cognitive screening as part of your Annual Wellness Visit, at no out-of-pocket cost. This is a brief assessment your provider performs to check for signs of cognitive impairment, including early dementia. If your doctor identifies concerns, Medicare Part B covers a more thorough cognitive assessment and care plan. After meeting the 2026 Part B deductible of $283, you pay 20 percent coinsurance for that deeper evaluation.
If you haven't had your Annual Wellness Visit this year, schedule one. It's free. Ask specifically about the cognitive screening component. Not every provider performs it thoroughly unless you request it, a lesson I've discussed before about being your own advocate in the exam room.
A screening won't tell you whether brain training is working. But it establishes a baseline. And a baseline is the only honest way to measure change over time.
A Gentler Way to Think About This
On my walk this morning along the trail past the Craggy Gardens overlook, fog sitting in the valley like something poured, I thought about Ernestine and her iPad. About James and his crossword. About the 2,802 people who signed up for the ACTIVE study in 1998, not knowing their data would still be teaching us something twenty-seven years later.
What keeps coming back to me is this: the brain wants to be used. It wants challenge and connection and rest and movement and surprise. In this season of life, it does not particularly care whether the challenge comes from an app or a conversation with a friend or learning to grow tomatoes from seed. The ACTIVE study found something real — one specific kind of training, reinforced over time, associated with meaningful protection. That matters. But it does not mean the brain is a muscle you can only strengthen in a gym designed by a tech company.
We are whole people. We have always been whole people. You don’t have to hold all of this by yourself.
Where to Start This Week
If the ACTIVE study results interest you, download BrainHQ and try the free exercises. Twenty minutes, three times a week. See how it feels. Don't expect miracles. Expect practice.
Then do the other things. Walk. Eat a real meal with someone you like. Get your hearing checked. Call your brother. Schedule your Annual Wellness Visit if it's been more than twelve months!
The best thing you can do for your brain this week probably doesn't require a screen at all. It requires showing up — for your body, for your people, for the ordinary, undownloadable work of staying alive to your own life.
May that be enough. I believe it is.
